Heat Management

[more_vampires]

It won't be cold enough here for another few weeks and although I appreciate your in depth answers, I should of said I was looking for you to use the SWAG method. Scientific Wild *** Guess - LOL. Seriously, what would you guess, I have a time in my head and I want to see what yours is and then I'll let you know what my unscientific results are once it gets cold enough.

also I posted something for you and anyone else to answer in my Lumen / Lux thread.

thanks

SWAG? Heh. Okay, straight from the air with no numbers. I'd say you might get 10-15% more turbo runtime before thermal sensor stepdown.

 

[Bigwilly]

SWAG? Heh. Okay, straight from the air with no numbers. I'd say you might get 10-15% more turbo runtime before thermal sensor stepdown.

So that equals about 30 more seconds. Interesting, I was thinking 20-25% more. I appreciate your answer and I'm curious to see what happens once the temperature drops. Thanks as always for your insight.

 

[more_vampires]

You could try it in a walk in refrigerator/freezer just for giggles.

Go to the convenience store and run it in the beer cooler or put it in the ice cream rack and close the clear lid.

"What are you doing?"
"Testing a flashlight, what does it look like?"

 

[Midnight.Sun]

...

Go to the convenience store and run it in the beer cooler or put it in the ice cream rack and close the clear lid.

"What are you doing?"
"Testing a flashlight, what does it look like?"

"..!!??"

 

[Bigwilly]

You could try it in a walk in refrigerator/freezer just for giggles.

Go to the convenience store and run it in the beer cooler or put it in the ice cream rack and close the clear lid.

"What are you doing?"
"Testing a flashlight, what does it look like?"

Lmao. That's freaking hilarious.

 

[more_vampires]

Glad I pulled a funny, guys! Cheered me up, too.

 

[WarRaven]

One small part of me looks forward to minus 30c temps,...
New lights!
Can't wait to honestly try ATR in the real outdoors here.

 

[Bigwilly]

One small part of me looks forward to minus 30c temps,...
New lights!
Can't wait to honestly try ATR in the real outdoors here.

-30c must play hell on your batteries. That's polar bear and penguin weather. Lol. Seriously, do you see a large drop in battery performance in such cold temperatures?

 

[WarRaven]

-30c must play hell on your batteries. That's polar bear and penguin weather. Lol. Seriously, do you see a large drop in battery performance in such cold temperatures?

Honestly, last three or four winters I've been using NiMH lights.
I had a couple old Surefire, ran them on AW RCR123, got maybe ten to twenty minutes run time, not much better on primaries.
Mind you, that was years ago.

I've no cold weather experience with 18650 and their lights yet, but old man winter is coming and darn straight I'll be posting my findings.
Though, to answer your question, yes I have saw lights just shut off in cold in my usage.
Until beginning of this year I used to drive a tow truck for a big company doing emergency roadside recoveries.
Saw all sorts of flashlight casualties, lost one too, so I carried three minimum on me.
Yes, I'm used to cold weather.

 

[Bigwilly]

Honestly, last three or four winters I've been using NiMH lights.
I had a couple old Surefire, ran them on AW RCR123, got maybe ten to twenty minutes run time, not much better on primaries.
Mind you, that was years ago.

I've no cold weather experience with 18650 and their lights yet, but old man winter is coming and darn straight I'll be posting my findings.
Though, to answer your question, yes I have saw lights just shut off in cold in my usage.
Until beginning of this year I used to drive a tow truck for a big company doing emergency roadside recoveries.
Saw all sorts of flashlight casualties, lost one too, so I carried three minimum on me.
Yes, I'm used to cold weather.

Is NiHM less susceptible to the cold? Have you thought about wrapping a light in some sort of thermal insulation so once you go out side they retain as much heat as possible?

 

[WarRaven]

Thanks for the thoughts, I no longer drive wrecker but still spend most of day outdoors of house anyhow.

The NiMH lights were charged daily an used hard, they got dimmer some nights then I wanted but kept working with me.
At the time I only had NiMH lights I trusted so I never thought of dragging Surefires out an trying that, run times were not acceptable even when used in summer but was high power guilt free then.

This winter I'm going to play around with wrapping materials, depending on light, I carry inside jacket but once it's out, its out in open with me and I'll get an idea of run times an outputs compared to my summer run times, I'm expecting about half but hoping lights internal heat will help offset cold some.
Then ATR should be a win-win, self heating an cooling ☺

 

[jmwking]

Is NiHM less susceptible to the cold? Have you thought about wrapping a light in some sort of thermal insulation so once you go out side they retain as much heat as possible?

Gloves?



-jk

 

[SemiMan]

As light outputs keep going up and size keeps going down at some point better thermal management is going to be needed. Some of the modded lights only get a few minutes of runtime on their highest setting. I've searched but haven't found much of what is being done to help with thermal management. At the cutting edge, I would think active cooling would be on the horizon. I asked one well known modder and he told me active cooling and waterproofing don't mix yet. Are any manufacturers working on this? Any other ways for managing heat? I thought I saw a post where a member had a light that had a modded computer liquid cooling system in the light but it didn't go into much detail. Are there any other lights out there like this? I just see a light small enough for a belt holster that puts out 14k lumens but only for about 3 minutes and can't help but think if some kind of thermal management would increase it to say 20 minutes would make the light so much more usable.

Actually, in many cases, as the light output has been going up and up and up, the problem has not been getting a whole lot worse, at least at the LED level.

When I was doing:
- 1000 lumens at 50 lumens/watt, I was dumping 17 watts of heat almost.
- Then at 100 lumens/watt, 7 watts of heat.
- Now at 200 lumens/watt, I am dumping < 2 watts.

(All figures rough ... and all LED lumens without optical losses).

One of the "problems" we have, is often we are pushing for maximum through, so even though we have a LED capable of 200 lumens/watt, we are running it at 100 lumens/watt for best throw.

So, I can now do 10000 lumens with the same LED heat load as 1000 lumens at 50 lumens/watt. When we start topping out with current LED tech in the 250 - 270 range, I could be into an LED heat load of 5 watts .... quite manageable.

Let's do a practical example of 10000 lumens at 150 LPW = 66 watts electrical, and about 33 watts of heat load at the LED. 150 LPW ----- I am going to use LED and system level efficiency interchangeably for ease of calculations.

The LED of course is not the only source of heat. I also have the electronics. Let's say I am doing 10000 lumens at a "practical" 150 lpw, or about 66watts. Now we are running into aspects of practicality and cost (and designer knowledge). If I start with a "reasonable" 12V battery, then hitting say 93% efficiency is pretty easy for this light, which means 0.07 * 66 = 4.7watts .... which is a reasonable amount to get rid of. Now let's say I take cost of the electronics out of the equation, assume I have the space for the right components, and get that conversion efficiency up to 97.5%. Now I am only getting rid of 1.65W ... again, quite manageable. I can use thermal compound to get the electronics heat out to the case for maximum cooling.

I have two other sources of heat. One, optical losses, lets say 10% to start or about 1000 lumens @ 300 lumens/watt = 3 watts. I may be able to use special coatings to get that down to lower, but 3 is a good starting number.

The other source of heat is the battery. Let's say I have 100mohm cells, 3 in series, 300mohm. 12V, 66W = 5 amps roughly, or 1.5W. I am going to double up my batteries and get that down to 0.75W. We also need to consider as the batteries age, that could creep up to 2W.

So 10000 lumens, 150 LPW ----- I am going to use LED and system level interchangeably for ease of calculations

66 watts
- 33 is heat of the LED
- 3 watts optical losses (Could be 2-5)
- 2 watts electronics (they are good electronics) (But could be 1.5-10)
- 2 watts batteries (But could be 1-5 or more)

These are all ranges and your mileage may vary, but it does give you an idea of what is possible.

Now when we are at 250 lumens watt, we now have 40 watts from the batteries, and the figures start to look like:

- 7-8 watts at the LED
- 2-5 at the optics (does not change)
- 1.2 - 7 at the electronics (scales linearly)
- 0.6 - 3 at the battery (scales linearly -- but these are likely to get better too).


Advanced in LED tech will significantly reduce the heat issue.

The other advancement in LED technology is high temperature operation. Hence you could have an "isolated" head that ran hot and dissipated a lot of heat, while the batteries and electronics were allowed to stay cool.

 

[Bigwilly]

I still think some form of active cooling is what is needed as the power of our flashlights goes up.

 

[MidnightDistortions]

I found that the best way to avoid the thermal limiter on the OR/Torchlab V.4 XPG Triple was to take the flashlight on a walk in below-freezing temperatures. Could be a challenging solution in some regions though.

Would be great to have as a hand warmer

 

[Bigwilly]

Would be great to have as a hand warmer

Then you have a reason to buy 2 of them. Lol.

 

[Gunner12]

All the heatsinking isn't going to help if there are voids in the LED or in the LED solder connections. Came a cross a study on LED stability today talking about voids and seem relevant here.
Here's a study by Cree:
http://www.cree.com/~/media/Files/C...pplication Notes/Solder Joint Reliability.pdf

and Osram:
http://www.osram-os.com/Graphics/XPic0/00091433_0.pdf/Processing of SMD LEDs.pdf

(The studies, goes into a lot more depth then I wanted to read, but the pictures show the issue well)

If we could get more efficient LEDs... then we get less heat!
What if you carried the light in a clear case/bottle, and filled that case with water? The bottom of the case would have to be flat as not to distort the beam. If the case starts to get hot, just change the water.

 

[Bigwilly]

All the heatsinking isn't going to help if there are voids in the LED or in the LED solder connections. Came a cross a study on LED stability today talking about voids and seem relevant here.
Here's a study by Cree:
http://www.cree.com/~/media/Files/C...pplication Notes/Solder Joint Reliability.pdf

and Osram:
http://www.osram-os.com/Graphics/XPic0/00091433_0.pdf/Processing of SMD LEDs.pdf

(The studies, goes into a lot more depth then I wanted to read, but the pictures show the issue well)

If we could get more efficient LEDs... then we get less heat!
What if you carried the light in a clear case/bottle, and filled that case with water? The bottom of the case would have to be flat as not to distort the beam. If the case starts to get hot, just change the water.

Good thought but water weighs over 8 pounds per gallon so I think the case would be too heavy.

 

[more_vampires]

All the heatsinking isn't going to help if there are voids in the LED or in the LED solder connections. Came a cross a study on LED stability today talking about voids and seem relevant here.
Here's a study by Cree:
http://www.cree.com/~/media/Files/C...pplication Notes/Solder Joint Reliability.pdf

and Osram:
http://www.osram-os.com/Graphics/XPic0/00091433_0.pdf/Processing of SMD LEDs.pdf

(The studies, goes into a lot more depth then I wanted to read, but the pictures show the issue well)

If we could get more efficient LEDs... then we get less heat!
What if you carried the light in a clear case/bottle, and filled that case with water? The bottom of the case would have to be flat as not to distort the beam. If the case starts to get hot, just change the water.

Reading now, thanks for the links. I love reading stuff like this!

Edit: MAN is that some science! Awesome. Looks like we're looking at 600 to 3000 cycles before a good chance of failure when a super high-drain light is started from -40 degrees. Cold solder joints (>50% void) result in failure.

From this, I was reinforced one more time that lead-free is garbage for electronics.

 

[more_vampires]

Good thought but water weighs over 8 pounds per gallon so I think the case would be too heavy.

Yeah, with that much weight just carry a travel cooler of beer. Occasionally throw the light into it and maybe get a brew while you're in there.